Suppose you have data on patients that record their height (in millimetres) and the length of their eyelashes (in millimetres). The aim is to perform clustering on this data. Explain why clustering using a distance measure such as Euclidean distance would be problematic on this data and outline how you could modify the data to address this.